how to build programming logic?

[bobcat2000]

Whhaaa!!!  Flowchart you say?   

You are lucky to see a single line of comment in the codes nowadays.  No CRC. No chart of any kind.  No Programming Standard.  No specs written.

Companies subcontract the programming oversea.  You try to tell the guy what you want to do and hope he can understand.  The codes come back and you are like WTF.

The last thing and the only thing I saw was a very big ER diagram from a big corporation.  The IT manager printed them all out and taped the printouts like wall paper covering the 4 walls of the entire office.  For the other companies, they just want to get their things done.  No chart no note no nothing.  They don't care.

So, go take some upper division classes from school.  The classes will teach you a lot of techniques to improve your programming logic.

[SiliconWizard]

In the old days - like, up to the 90s or so - people were actually analyzing before programming. That was actually a real  job. Nowadays, you don't, and specifying, analyzing and programming are pretty much all mixed up (except in niche fields - safety critical stuff - and even so, specifying is then a separate job indeed, but "analyzing" vs. "programming", per se, is not really anymore.)

Thank cost reduction and the agile fad. Writing specs has become an insult. Analyzing, defining architectures before coding? You bet your ass not. It's so inflexible and so unproductive (so they claim.) You code, code, code away, write lame tests with pieces of your own code too, and move on (so basically, your tests prove that your code is your code is your code.) Who cares anymore about programming logic? If you don't know how to do something, just use a library. If you're not even quite sure what it is your code is supposed to do, you don't really care. There are no specs.

[MarkMLl]

Does anybody use flowcharts anymore?  They were a big deal in the early '70s when I started programming and could be quite detailed.  IBM documentation often included machine printed flowcharts.

I still scribble them down from time to time depending on the complexity of the function I am writing.

Apart from what other people are saying, they were largely obsoleted by Jackson Structured Programming, Unified Modelling Language and the rest. And even when they were popular there were alternatives, e.g. one promoted by Dijkstra.

I was dealing with a particularly complex startup sequence in some embedded code in about '93, and drew a flowchart to get to grips with it. And even then I felt I was doing something decidedly... well, I'd have said "retro" but I don't think the word existed in those days :-)

MarkMLl

[rstofer]

One of the CS luminaries once declared that programmers shouldn't have access to a keyboard.  The layers have changed and it's probably software designers who shouldn't have keyboards.

From my FORTRAN days I can understand #652.  The best FORTRAN programs were created after the programmer dropped the box of cards.  Today we probably call it 'refactoring'.

https://medium.com/@caldavis/1256-quotes-about-technology-curated-fbe7f8e791cb

[PlainName]

Does anybody use flowcharts anymore?

Kind of. I pretty much always do a state transition diagram for a state machine. Flow charts, not so much but if it makes a non-obvious thing simpler then I will. The thing nowadays is UML (or SYSML for embedded) but you really need to speak it to read it, and a flow chart is understandable without prior training.

[SiliconWizard]

Yeah. The UML spec has grown to almost 800 pages.

[MarkMLl]

One of the CS luminaries once declared that programmers shouldn't have access to a keyboard.

Can you provide a reference for that? It has the feel of Fred Brooks about it, and many of the things that he said (in The Mythical Man-Month etc.) became perfectly obvious when his rationale was considered.

Without further explanation, I'd suggest that it might parallel the productivity gain seen inside Sun (I think) when Powerpoint presentations were banned: if a programmer doesn't have access to a keyboard then he won't be tempted to tinker endlessly.

MarkMLl

[rstofer]

One of the CS luminaries once declared that programmers shouldn't have access to a keyboard.

Can you provide a reference for that? It has the feel of Fred Brooks about it, and many of the things that he said (in The Mythical Man-Month etc.) became perfectly obvious when his rationale was considered.

Without further explanation, I'd suggest that it might parallel the productivity gain seen inside Sun (I think) when Powerpoint presentations were banned: if a programmer doesn't have access to a keyboard then he won't be tempted to tinker endlessly.

MarkMLl

I can't remember the author, maybe Dijkstra or Wirth, somebody of that stature.  I haven't seen the quote in more than 20 years and the constext was that programmers should be designing their code with paper and pencil and leaving typing to someone else.

In the early days, we had an entire department full of keypunch operators.  I could never use their services due to budget constraints but I still wrote my code on Fortran Coding Forms - then I keypunched it.

[Nominal Animal]

If OP was asking career advice, then I'd have recommended self-expression/assertiveness/speaking/image projection course with a professional.

In today's business world, what you actually can do is secondary; the optics are primary.  Dress appropriately (slightly smarter than expected, but without making it obvious), speak the lingo, project an image, and be constantly on the lookout for switching jobs upwards.  Do not stay in one place for longer than three years, absolute maximum; preferably switch jobs just below the two year mark.  Be a people person, but never take any risks in personal relationships: they can backfire, and ruin your career.  Even innocuous help can backfire if the person does something inappropriate, so don't do it.

Be ruthless.  Never explain or admit any errors, just move on.  Take all glory from successes you participated in; which also means you need to choose what you participate in based on their likelihood of success, not on any innate interest.

Personally, I just can't do that shit.  I've chosen to be poor but happy and non-broken.

[brucehoult]

Personally, I just can't do that shit.  I've chosen to be poor but happy and non-broken.

I was getting worried there for a moment!

Me too.

At least the interesting jobs started to pay better once I got past 45, and I also got to live in or regularly visit interesting foreign places such as Moscow and San Francisco. That seems to have helped get better offers from NZ companies afterwards too -- my disposable salary after taxes, deductions, rent, car insurance, utilities is now better working for a NZ company than it was at SiFive in San Francisco, though not quite as good as at Samsung in Moscow (slightly lower gross salary but low taxes, low rents, nearly free utilities).

Retirement is a worry though. I guess I never will.

[MarkMLl]

I can't remember the author, maybe Dijkstra or Wirth, somebody of that stature.  I haven't seen the quote in more than 20 years and the constext was that programmers should be designing their code with paper and pencil and leaving typing to someone else.

In the early days, we had an entire department full of keypunch operators.  I could never use their services due to budget constraints but I still wrote my code on Fortran Coding Forms - then I keypunched it.

Wouldn't have been Wirth, he was a hands-on guy. If it were Dijkstra he needs to be taken in context: he was inclined to be /intentionally/ stroppy (his most popular quotes come from his "how do we tell truths that hurt?" paper), and he developed a massive grudge against IBM for- he believed- ignoring the significance of his work.

(You had coding forms? Luxury! We had to toggle in the loader by hand, or load tests from paper tape with the consistency of airgun target.)

MarkMLl

[MarkMLl]

I can't remember the author, maybe Dijkstra or Wirth, somebody of that stature.  I haven't seen the quote in more than 20 years and the constext was that programmers should be designing their code with paper and pencil and leaving typing to someone else.

Wouldn't have been Wirth, he was a hands-on guy. If it were Dijkstra he needs to be taken in context: he was inclined to be /intentionally/ stroppy (his most popular quotes come from his "how do we tell truths that hurt?" paper), and he developed a massive grudge against IBM for- he believed- ignoring the significance of his work.

Apropos "a programmer should not use a keyboard": I had a few minutes while the coffee machine came to life...

I think it was an apocryphal story /about/ Dijkstra, based on an apocryphal (and now disproven) story that Mozart composed without a keyboard.

It is contradicted by EWD 279 https://www.cs.utexas.edu/users/EWD/transcriptions/EWD02xx/EWD279.html where he writes

As target system I am envisaging a "programming laboratory", as I seem to need this for the education of software engineers.
...
The kind of hardware I am looking for is a simple, very fast machine, with a reasonably sized primary store, a very large backing store, a line printer, a few keyboard terminals
...
The keyboard terminals are viewed as "the programmer's desk". I need at least two of them, to create the environment of parallel interactions (extensions or modifications) with the same program, perhaps an "operators console" in addition to that. If the laboratory grows towards an environment in which larger numbers of students can get experience, more might be needed for reasons of more intense usage of the system.

Mention of an operator's console in that context does not mean that there was a dedicated system operator present. The sort of machines with which Dijkstra was familiar might have supported interactive terminals, but the operation of starting up (and if necessary killing) the interactive subsystem could not necessarily be handled by the terminal itself... and very often a console was needed to e.g. tell the system that a scratch tape had been mounted or in extremis as output for a system memory dump.

I also found this:

Apart from the two telephones, another notable feature of the office in Dijkstra’s home was an elegant portable Olivetti typewriter. At the time, a proper professor would write with a fountain pen, possibly a Mont Blanc, which was at the time one of several ordinary brands. Only secretaries touched a keyboard. But Edsger typed his EWD memo’s himself, somewhere acknowledging the unversity for purchasing for his sole use the one typewriter with the one font that could do justice to his thoughts. This way he did a lot of typing: the big one, “Notes on Structured Programming” (EWD249), runs to 88 pages.

Fast forward to my visit to Edsger in the 1980s in Austin, Texas, when all profs did their writing on keyboards and Edsger refused to touch one. I have since been told that one way of swearing allegiance to Edsgerismus is to purchase a Mont Blanc and to find that it facilitates the proofs of your programs (which are only to be contemplated, not executed). In his office Edsger had a pencil dangling from a string with a sign pointing to it saying “Word Processor”.

Not only were word processors anathema to Edsger as superfluous mechanical devices, but they also gave offense by coming with fat manuals. Edsger told me indignantly that some manual for Wordperfect ran to six hundred and fifty pages. He reminded me of the fact that Georg Joos only needed six hundred for his great work “Textbook of Theoretical Physics”. It was blasphemy to Edsger that a manual for buggy software for a perverse purpose used more space than a comprehensive compilation of the most sublime knowledge.

https://vanemden.wordpress.com/2008/05/06/i-remember-edsger-dijkstra-1930-2002/

MarkMLl

[rstofer]

Interesting history!

When I started using the IBM1130 in 1970, we had 2 machines in an 'open shop'.  Walk up, take complete control of the machine.  There were no operators for these machines other than the users - primarily aerospace engineers (except me...)  Oddly, the keyboard and typewriter weren't used for much beyond playing 3D Tic-Tac-Toe (Robert Louden's IBM 1130 book had the code).

https://en.wikipedia.org/wiki/IBM_1130

In the same room was a CDC 1700 which was a remote terminal for the CDC 6400 up at the mother plant.  That machine had an operator.  Submit your card deck, come back later for the results.

Grad school (circa'75) was a DataGeneral <something> with teletype machines around the campus.  I had dial-up access with a 110 baud acoustic coupler that I eventually upgraded to 300 baud modem.  I had a lot of fun playing Startrek on that machine.  Writing a Tiny Algol compiler in Basic wasn't easy...

There's a reason I built an FPGA incantation of the IBM 1130 that runs all of the factory software, unchanged.

[MarkMLl]

Ah yes, the IBM 1130. I ended up writing my own 2741 emulator so I could play around with its APL implementation... then went on to modify it to be useful in a whole lot of other contexts. But since the character encoding etc. was determined by the design of the Selectric mechanism, things like cut-and-paste became /exceptionally/ hairy. I'm not sure whether one should be sneakily proud of something which is so difficult to maintain :-)

In actual fact, the 1130 and competing machines like the LGP-30 (and the Burroughs L and TC electromechanical accounting machines, and a whole lot of other stuff) are part of a "personal computer" tradition that reaches back to ENIAC via Stan Frankel. Arguably, IBM's APL successfully tapped into that tradition: which of course has links to both Wirth (who supervised Breed's early implementation at Stanford) and Dijkstra (who publicly derided both APL and IBM).

MarkMLl

[SiliconWizard]

One of the CS luminaries once declared that programmers shouldn't have access to a keyboard.

Can you provide a reference for that? It has the feel of Fred Brooks about it, and many of the things that he said (in The Mythical Man-Month etc.) became perfectly obvious when his rationale was considered.

Without further explanation, I'd suggest that it might parallel the productivity gain seen inside Sun (I think) when Powerpoint presentations were banned: if a programmer doesn't have access to a keyboard then he won't be tempted to tinker endlessly.

MarkMLl

I can't remember the author, maybe Dijkstra or Wirth, somebody of that stature.  I haven't seen the quote in more than 20 years and the constext was that programmers should be designing their code with paper and pencil and leaving typing to someone else.

Yep. I remember this quote too, and can't remember who either. This is congruent with what I said earlier though, that analyzing was a job in itself and separate from programming, and that this was considered good practice.

But even if the same person does both activities, I still think it's good practice, and this is, among other points, why I find trying to fight the alleged "verbosity" of programming languages completely irrelevant - something that most modern programming languages have tried to do, and that a majority (or so it seems) of developers are on board with. As though saving a few keystrokes would make a difference. If you actually design your programs before coding them, it absolutely doesn't. Of course, if you start typing away frantically without any kind of prior analysis, I suppose it would.

[rstofer]

Yep. I remember this quote too, and can't remember who either. This is congruent with what I said earlier though, that analyzing was a job in itself and separate from programming, and that this was considered good practice.

But even if the same person does both activities, I still think it's good practice, and this is, among other points, why I find trying to fight the alleged "verbosity" of programming languages completely irrelevant - something that most modern programming languages have tried to do, and that a majority (or so it seems) of developers are on board with. As though saving a few keystrokes would make a difference. If you actually design your programs before coding them, it absolutely doesn't. Of course, if you start typing away frantically without any kind of prior analysis, I suppose it would.

The Bureau of Labor Statistics differentiates programmers from developers

https://www.bls.gov/ooh/computer-and-information-technology/computer-programmers.htm
https://www.bls.gov/ooh/computer-and-information-technology/software-developers.htm

There's quite a difference in pay scale as well as scope of work.

[brucehoult]

There's a reason I built an FPGA incantation of the IBM 1130 that runs all of the factory software, unchanged.

In my first year at university (1981) the computer lab for 1st year students had a PDP 11/34 with 256k RAM, 22 VT100 terminals, and 2 LA120 printers.

The year before us (I'd visited a friend who skipped 7th form) they'd used it with an environment where you edited FORTRAN then when you said to RUN it you were forcibly logged off (so someone else could have a go) and you went to the printers to get the result of your compilation and/or run. If it didn't work then you had to fight for a terminal and have another go.

But our year got a different environment. The machine was running a light weight multi user OS (RT11? Not RSTS anyway) with a shell that gave you a few basic DIR, TYPE, PRINT, RENAME, DELETE commands (I think there was *both* COPY and PIP) and a locally-written line editor. The editor code was shared between all users, and you could edit I think a maximum 8k text file (1 MMU segment). Pascal compilation was provided via the NBS (National Bureau of Standards!) Pascal compiler running in a batch queue, so only one compile could actually happen at a time. Later in the year this was changed to the more powerful but slower OMSI Pascal. Once you'd got your program compiled you were free to keep the binary and run it as and when you wanted (subjected to VERY tight disk quota -- you had to reduce to maybe 32 KB before you could log off)

One of the first things I did was write a light weight fell-screen text editor so I didn't have to use the provided line editor. As I recall, I had a doubly-linked list of lines of text, with each line a singly-linked list of records with an array of 6 chars and a (2 byte) pointer to the next chunk. And a free-list of line and char chunk records. The line with the cursor on was copied to an 80 char array. I only implemented typing to insert (including newline), backspace to delete, and arrow keys to move. No copy and paste -- if you wanted large scale rearrangements you could quit and use the provided line editor. It was still a huge improvement over the line editor. I left a world-execute copy in my home directory and told friends about it. I don't remember the executable size now but it was tiny -- and of course shared between multiple people running it.

Oops .. that was a sidetrack. There was a dusty old IBM 1130 in the back corner of the terminal room. I never saw it running. It featured a couple of times in lectures, primarily about the 28 pass FORTRAN compiler. Sad that I didn't get a chance to try it, if it even still worked.

[MarkMLl]

In my first year at university (1981) the computer lab for 1st year students had a PDP 11/34 with 256k RAM, 22 VT100 terminals, and 2 LA120 printers.

When I went up in '75 we were expected to use card punches for FORTRAN... there was another bank used by maths types for some variant of ALGOL and Teletypes along one wall for the bearded postgrads. By the time I left the cardpunches had been replaced by some variety of terminal, but I believe the overall job cycle was still very similar: enter your program, run it, and come back later for the printout.

One of my lecturers was an engaging chap by the name of David Brailsford :-)

Subsequent to that I worked for a mainframe company which was an interesting experience: they were heavily into automating bank and building society branches with (by our standards) ridiculously underpowered computers giving good response on hundreds of terminals. And I spent a few years at a university where the central computer system was Multics, and there was a jury-rigged system of Kermit and Gandalf boxes hooking things together campus-wide.

MarkMLl

[rstofer]

I guess you all realize that we have lived through exciting times.  And continue to do so...
Can you even imagine what the next 50 years will bring?

[MarkMLl]

I guess you all realize that we have lived through exciting times.  And continue to do so...
Can you even imagine what the next 50 years will bring?

Yes, "The Limits to Growth" was required reading for one of my courses.

MarkMLl

[brucehoult]

I guess you all realize that we have lived through exciting times.  And continue to do so...
Can you even imagine what the next 50 years will bring?

I'm hoping I won't have to imagine the first 1/2 to 2/3 of that!

[brucehoult]

I guess you all realize that we have lived through exciting times.  And continue to do so...
Can you even imagine what the next 50 years will bring?

Yes, "The Limits to Growth" was required reading for one of my courses.

Yeah, we had to read that too. And "Future Shock". What total bullshit they both were! The number of times we've supposedly been about to run out of this or that in 5 or 10 years -- starting 40 years ago!

Much better to read Pournelle's "A Step Farther Out" from 1979.

[SiliconWizard]

Uh. I've never read Future Shock, but from the description of it ( https://en.wikipedia.org/wiki/Future_Shock ), it seems not that far off. Maybe a few "predictions" were bogus (as most predictions regarding resource depletion are), but I think it still made a number of pretty valid points. Probably too "pessimistic" about dates - things didn't happen quite as fast as it depicts - but we're not that far off from what seems to be described there now. Again, I only read a summary, so maybe there were also many bullshit points in it. But also, maybe the way you read it while being a student was different from how you would perceive it these days?

[brucehoult]

I don't think the wikipedia article captures the flavour of the book.

Yes, most of the concrete predictions listed in wikipedia have happened. That's not, I think, the point. Everyone from Clarke to Pournelle was predicting those things.

The difference with Toffler, and why I thought he was wrong then and still think he is wrong, is that he regarded these predictable changes as A BAD THING. He regarded it all as being a disaster for human mental health, with people overwhelmed and without hope with no firm foundations. No permanent place. No permanent relationships. etc.

And, on top of that, he was one of those predicting the imminent exhaustion of metals (by 1980 if I recall), oil and gas and coal, constantly increasing pollution, and so forth.

It's all just so wrong, and for very fundamental reasons.

The peak of pollution was just after WW2 and in the 1950s. Since then everything from the rivers and oceans to the air in cities (and globally) has become far *less* polluted. Except in places where the government is in charge of everything, such as the Soviet Union and today's China.

Toffler totally missed that while people might change their physical address and phone number often, other means were developed that enable maintaining permanent relationships -- phone number portability (at least within a country), permanent email addresses (I've been bruce@hoult.org for more than 20 years and always will be in future), and semi-permanent accounts on facebook, twitter, skype, and dozens of others including eevblog -- any one might disappear due to accident or some organisation ceasing to exist, but there is redundancy. In any event they tend to last a lot longer than jobs, phone numbers, and physical addresses.

It must be incredibly difficult for a young person now to imagine how isolating it was to be a teenager (or a wife) on a farm in a rural area in the 1960s. A rural area anywhere even in the UK or USA, but all the more so in somewhere like New Zealand or Australia. There was maybe radio and a newspaper and gradually TV to keep track of what was happening in the outside world, but that was very filtered and superficial. You could get more specialised magazines such as Scientific American or Wireless World, but those were literally three or four months out of date when you got them.

The worst part was that you didn't have a voice. You couldn't *participate*.

Now I can choose to live at a remote beach in the most isolated region of one of the most isolated countries in the world (New Zealand) but I can participate fully (especially in these times without physical conferences) in any world-wide interest I might have. For example I can participate in RISC-V task groups designing new instruction set extensions, and have an actual influence over the results. I can contribute to LLVM or the Linux kernel or anything else that attracts my interest. I can be EMPLOYED by a company anywhere in the world. I can communicate instantly -- by written word, photos, voice, or video -- with anyone who I choose to stay in contact with.

The world -- fifty years after Toffler (and the Club of Rome, and others) made dire predictions of societal and physical collapse 10 or 20 years out -- is not a disaster, it it INFINITELY better in every way than the world of 1970.

[SiliconWizard]

I get your points, but many of them are definitely debatable.